Name
Abstract | ||
|---|---|---|
This paper presents a distributed acoustic sensing system based on direct detection phase-OTDR (optical time domain reflectometry) technique along with a deep learning based threat classification approach. Signal needs to be processed with denosing and signal conditioning algorithms prior to threat classification. For threat detection, power thresholding approach is taken. The developed system and algorithms are tested experimentally using a buried fiber optic cable for distances up to 40 kilometers. The results show that by using appropriate signal conditioning and threat detection algorithms, six different activities such as manual digging and walking/running can be classified at 40 kilometers distance and up to 10 meters away from the fiber optic cable. |
Year | Venue | Keywords |
|---|---|---|
2017 | Signal Processing and Communications Applications Conference | Distributed acoustic sensing,phase-OTDR,deep learning,convolutional neural networks,CNN,threat detection,threat classification |
Field | DocType | ISSN |
Time domain,Signal conditioning,Optical fiber,Computer vision,Computer science,Distributed acoustic sensing,Artificial intelligence,Reflectometry,Thresholding,Deep learning | Conference | 2165-0608 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Aktas, Metin | 1 | 4 | 3.44 |
| Toygar Akgun | 2 | 90 | 9.39 |
| Umut Demirçin | 3 | 41 | 4.20 |
| Buyukaydin, Duygu | 4 | 2 | 1.06 |